Guest Post by Willis Eschenbach
Anthony has posted on a recent study (behind a paywall) of rice production in Rice yields, CO2 and temperature – you write the article. The article claims that rice yields are falling, and will fall further, due to temperature changes.
I said I’d write the article if someone would send me the actual study, and a couple of WUWT readers came through, my thanks to them. Here’s my take on what the authors have done.
First, the good. They have used actual data from farmer’s fields, rather than theoretical data or greenhouse experiments.
Figure 1. Rice Fields in Ha Giang
However, there are some troubling things in the study.
First, it covers a very short time span. The longest farm yield datasets used are only six years long (1994-99). Almost a fifth of the datasets are three years or less, and the Chinese data (6% of the total data) only cover two years (1998-1999).
Now, if they were comparing the datasets to temperature records for the area where the farms are located, we could get useful information from even a two-year dataset. But they are not doing that. Instead, they say:
Data series from the weather stations at the sites were too short to determine trends. Instead, trends in Tmin and Tmax were based on a global analysis of ground-station data for 1979–2004 …
Unfortunately, they have neglected to say which “global analysis of ground-station data” they are using.
But whichever dataset they used, they are comparing a two year series of yields against a twenty-six year trend. I’m sorry, but I don’t care what the results of that comparison might be. There is no way to compare a two-year dataset with anything but the temperature records from that area for those two years. This is especially true given the known problems with the ground-station data. And it is doubly true when one of the two years (1998) is a year with a large El Niño.
For example, they give the trend for maximum temps in the winter (DecJanFeb) for the particular location in China (29.5N, 119.47E) as being 0.06°C per year, and the trend for spring (MarAprMay) as being 0.05°C per year (I get 0.05°/yr and 0.04°C/yr respectively, fairly close).
But from 1998 to 1999, the actual DJF change was +2.0°C, and the MAM change was minus 1.0°C (CRU TS Max Temperature dataset). As a result, they are comparing the Chinese results to a theoretical trend which has absolutely no relationship to what actually occurred on the ground.
To try to find out which “ground-station data” they used, I compared their temperatures for China (29.5N, 119.47E) with the CRUTEM3, CRU TS, and GISS records from KNMI. However, I could not match their numbers, although I could get sorta close. Worrisome.
Next, they have not mentioned autocorrelation anywhere in their study. This makes me think that they have not adjusted for autocorrelation, which is particularly important with short datasets.
Next, they base their predictions for the future on a single computer model of the regional changes. It is widely agreed that computer climate models are not very good at predicting regional changes, so that part of their study seems very weak.
I am most mystified by their use of the 26-year temperature trend. Why not use the actual year-by-year changes in the local temperature? The rice is responding to actual temperatures, not to a mathematical trend … so why not compare yields to actual temperatures?
So once again, we have questionable methods used with uncited data to give alarming results. It is too bad, because their premise is good, and so is their general approach (use actual farm data).
Color me unimpressed …
Mike {previous thread}says:
August 10, 2010 at 1:03 pm
“Published in the online early edition the week of Aug. 9, 2010 in Proceedings of the National Academy of Sciences —a peer-reviewed, scientific journal from the United States—the report analyzed….”
PNAS is not exactly peer reviewed in the generally accepted sence.{sic} If your buddy is a member of the national academy he or she can communicate it to PNAS and it will be published with nothing more than a spelling check.
———————————————-
If that is correct, maybe explains a lot!
While Mr. Welch could perhaps be cut some slack, surely some of the allied authors would have had some on the ground input.
BillD says:
August 11, 2010 at 4:52 pm
That would be an incorrect idea. No one is against peer-review, we’d like to make it actually work, or work better where it is working.
Also, I’m hitting the balls as fast as they are pitched. By that I mean I try to look over all of the articles in the field and find the ones that are obvious howlers. In addition I take on challenges like this one on rice. So my time is not long.
However, if you think that there is a good study on climate out there, send it to me in an email and I’ll take a look. You can reach me at willis [at) taunovobay.com, with the usual changes in the email to foil the email harvesting robots.
Thanks Willis,
Looking for background yesterday I found the following amazing study:
http://aob.oxfordjournals.org/cgi/content/full/95/4/695
• Methods Rice plants were grown outdoors in plastic pots and moved at heading time to three temperature-controlled glasshouses under high night temperature (HNT; 22/34 °C), high day temperature (HDT; 34/22 °C) and control conditions (CONT; 22/22 °C). Grains were sampled periodically, and the time-course of grain growth was divided into rate and duration by logistic regression analysis. Endosperm cell numbers and cell sizes were analysed by digitalized hand-tracing images of endosperm cross-sections.
I had to read a bit to understand that they reversed conditions for the plants, and HNT means at night 34C and during the day 22C.
Better than cutting off frog’s legs and checking height of jump, since rice cannot croak/scream.
It is amazing what is called science in this day and age.
Anyway, maybe this is the origin of the harm brought from high night temperatures :).
There seems to be hope though for the science and logic, in another abstract:
http://www.ingentaconnect.com/content/bsc/sspn/2009/00000055/00000001/art00014
The effects of night temperature on biomass accumulation and plant morphology were examined in rice (Oryza sativa L.) during vegetative growth. Plants were grown under three different night temperatures (17, 22 and 27°C) for 63 days. The day temperature was maintained at 27°C in all treatments. The final biomass of the plants was greatest in the plants grown at the highest night temperature. Total leaf area and tiller number were also the greatest in this treatment. Growth analysis indicated that the relative growth rate in the 27°C night-temperature treatment was maximal between days 21-42 and this was caused by increases in leaf area ratio, leaf weight ratio and specific leaf area. Plant total nitrogen contents did not differ among treatments. However, nitrogen allocation to the leaf blades was highest and the accumulation of sucrose and starch in the leaf blades and sheaths was the lowest in the 27°C night-temperature treatment by day 42. Despite this, dark respiration was also highest, and both the gross and net rates of CO2 uptake at the level of the whole plant at day 63 were the highest in the 27°C night-temperature treatment. Thus, high night temperature strongly stimulated the growth of leaf blades during the early stage of rice plant growth, leading to increased biomass during the vegetative stage of the rice plants. As the CO2 uptake rate per total leaf area was higher, photosynthesis at the level of the whole plant was also stimulated by a high night temperature.
Hey, the science is not settled :). Who would believe it!
A lot like the flyer I just retrieved from the letterbox advertising ‘ADELAIDE’S PSYCHIC EXPO’ this weekend. With such abundant access to thiotimoline why on earth would you want to kill trees and expend all that CO2?
It amazes me that a) someone would fund such a useless study and b) that anyone would publish such a study which provides no data of use. It seems that so long as it follows the official IPCC line of produce scare mongering papers, the quality of the work is regardless and it is published either way – this only seeks to undermine climate science in general and turn off the public. The public are not stupid, they known BS when they see it!
Its not that the scientists are failing to engage the public, they are publishing rubbish purely to create an alarmist headline for the media. They will bring about their own end, without the help of sceptics or “oil funded lobbyists”!
Just another typical example of climate ‘scientists’ publishing papers, where the standard of research would be rejected outright as being inadequate in any field of real science.
BillD: August 11, 2010 at 4:52 pm
I wish that just for a change that Anthony and WUWT could find a mainstream peer reviewed article on the environment that meets the standards and expectations of the readers of this web site. I am usually asked to peer review about 30 articles per year and the quality submitted manuscripts is actually quite high, although most papers benefit from critical reviews.
WUWT is a commentary on “the puzzling things in life, nature, science, weather, climate change, technology, and recent news…” Good, quality, peer-reviewed articles should be thought-provoking, but not puzzling.
And puzzling, rather than thought-provoking, is what’s been coming out of the “climate science” community for years…
This says everything. The Big Disconnect With Reality. Just like turning good individual station records into anonymous gridded soup.
Me too. Now I wonder if he was just learning ecneics so that he could more plausibly appear to support the BS. I read a very good account of Monica Lewinsky’s story (I like to see people in the round, not just all-bad or all-good) and one thing I remember is that in addition to Clinton’s charismatic, genuinely likeable qualities, he admitted lying, telling lies, as long as he could remember.
This is only part of the continuing warmist policy. Some years ago the late Prof. Schneider advocated “exaggerating” man-made climate fears in order to scare the public. “You have to find ways to exaggerate the threat”
Thank you, Willis, for this excellent and beautiful analysis.
I think it bears pointing out that your critique is based on looking at what these researchers actually said they did. It is not based on ‘this is a paper by proponents of AGW, so we don’t like it no matter what’ – something continuously bleated about by those AGW proponents. That is a very important difference.
And may I say that it warms the cockles of the heart of this old scientist to read your critique. My old professors would have done exactly as you did, had we handed in such a thesis. Only – they would not have been so kind but far more acerbic. And there would have been tears … oh yeah!
That study is something I consider sloppy science.
Can anyone answer a question about ‘teleconnection’ – the effect whereby the growth of a tree or rice plant supposedly reflects global temperature but doesn’t match the local temperature…
On the face of it, this sounds nutty. However, couldn’t the plant be responding to sunlight rather than temperature? Teleconnection then makes more sense if solar activity is affecting global temperature.
However, this is just a random thought, not based on any data. I was hoping someone more learned could comment on this suggestion.
A few articles back we spoke of University Administrators not being Researchers but just academic Politicians. Now we discuss yet another example of Peer Reviewed nonsense being published by “PNAS”, that just falls apart in your hands as you read it (thanks to Willis and others). Now I know I’m slower than Leif and a few hundred others here, but it just seems that there’s no one who knows anything that’s in control of anything anywhere and it looks like Fat Albert and his gang of misfit, global-warming, carpetbaggers and speaking-fee-collectors are running hollywood and the federal government and the united nations and every college and university, and every professional organization and publication on the planet.
Did I get that right? Did I leave anyone out?
I know this blog is for weather and climate scientists. But there are a lot of people out there you could serve and enlighten about the nonsense of AGW, IF you put your cookies on a lower shelf by explaining scientific principles and terms. I know one of the objectives of modern science seems to be to make things so complicated that everybody else is totally intimidated and in awe and worshipping the elitarian doctors and masters–That’s why they often talk in latin!–BUT this morning in our rice-producing country of Taiwan’s main AGW propagandiser English language shitsheet, “The China Post” carried this “research” as a scary harbinger article of AGW. I could not refute it, because I had to read this above twice to get the major drift. But if you would write in clearer terms (even explaining jargon in normal language in parenthesis like these) you might give normal people like me ammunition to refute the lies that get printed here, and most likely in other countries.
You may fight the scientific war in some ways, but you must see that they–the warmists–are winning the propaganda war with their big media guns, and unless you fight AGW (like I do on my blog) with stuff that people understand, you can kiss your and our future of truth in science goodbye. They will win!
They ARE winning, because the mainstream media are behind this globalist scheme to make a one world government. That is why I really admire Lord Monckton. He puts things more in terms lay people can understand and addresses normal folks.
I know that you are scientists, but you could simplify more and explain things better. The comments below the articles are sometimes easier to understand than the actual article. I love this website, don’t get me wrong, but if your objective is winning the information war, you are mainly preaching to the choir, if you don’t write in simpler terms and explain difficult concepts and techincal terms. How about an on-board Climate thesaurus widget to look things up?
Contrary to what a commenter wrote a little above this one, “The public are not stupid, they known BS when they see it!”, I find that the public reads newspapers and imbibes TV as Gospel! And that is why we have to change to reach out to the uninformed!
The study uses correlations between farm yield and ‘weather variables’ from stations within 15-20km of the farm to build a statistcal model. That statistical model is then used to estimate the effect of changes in temperature over the 26 year ‘global analysis’ record on yield. Your post seems to suggest that the authors used the 26 year record to come up with their statistical model for how rice yield responds to temperature. This would indeed be silly, but it is not what was done.
Additionally your post (and frankly the article so its probably not your fault) seems to indicate that the only thing used in the 26 year record as the ‘global value’ – i.e. they only cared about the global average. This is not true. They used the NCDC 5×5 degree gridded temperature analysis, and took the trend in the gridpoint containing each farm to estimate the effect of late 20th century climate change. This is a little hidden at the bottom of the ‘Data and Methods’ section.
Finally, maybe I missed it, but I can’t find any place in which they used a general circulation model to predict anything. The only model they use is a simple regression one, in which they control for whatever they can, and estimate the effect of maximum and minimum temperature as well as radiation. The only real prediction of the future they make is to note that Tmin is projected to rise more than Tmax, potentially skewing the net effect towards a decrease in yield. However, they make a point of noting that their results should not be interpreted as a prediction of decreasing yields by 2100. The results refer to the marginal effect of warming and “should not be extrapolated to the nonmarginal warming that is projected to occur in Asia by the end of the century.”
Oh Yeah. The global trend is what influences local growth. That is why we in Wyoming are awaiting our day to enter the banana trade.
Just use the local data, boys. If there is adequate variation in temperature, and if there is truly a significant effect, the data will show it. Have some faith.
Willis, I didn’t read all the comments so it may have been mentioned but you appear to have missed the real malfeasance in the Rice article. The claim was that global warming had reduced the “rate of growth” in unit rice production, not the rate of unit production. I don’t care what dynamic system we are discussing, ultimately, as we approach optimum conditions, the growth curve approaches a horizontal asymptote and therefore the growth rate slows . Temperature records or no, the paper should have been rejected for this reason alone. For example, this slowing in growth rate in unit production can also be related to my increase in age, or the rusting of the Brooklyn Bridge.
re: Lu says: August 12, 2010 at 7:52 am
Hi Lu,
Your comment is heartfelt, well written, and certainly brings up some legitimate points. I would like to try to address some of those issues.
The first point I would like to make is that this site actually isn’t at all for just weather and climate scientists. While there are some who frequent the site, I daresay the majority are not. Some of us are scientists in other fields, and a lot of folks aren’t scientists at all. A large goal of this site, as I see it at least, is very much to help educate folks about some of the basic precepts of science – even if the mode of that education is thru reading articles that use those exact principals to deconstruct some of the shoddy ‘science’ that is out there – or to evaluate other potential causes than those being commonly cited.
For non-scientists, it is difficult and a bit of a learning curve. The comments sections can be incredibly helpful as you noted yourself. A great use for them is for anyone who doesn’t understand a point to question it in the comments. Unfortunately you won’t always get a reply with a good solid answer – but more often than not you will and that’s far more than can be said for the majority of sites out there.
I, and I’m sure many others, would love to see a ‘glossary’ of sorts- the problem is that it takes a large amount of time and effort to put something like that together. Folks posting here are volunteering their time as it is, and doing it out of personal interest in the topic(s) on their free time outside of their normal jobs. What I’d LOVE to see, although it still takes a lot of time and effort on the part of a site owner and site moderators, is a joint effort where everyone contributes to creating that ‘glossary.’ In that fashion an incredible amount of information can be put together far easier than if just site owner and moderators were to try to do the entire project – but it would still require a lot of time on their parts just to double check submissions, correct errors, eliminate repeat entries, create the format & layout & webpage(s) for the final product, maintain that page, etc…. This from folks who are already giving up hours and hours of their own lives to this site as it is.
What probably could be done a bit easier is to provide links to some existing ‘science’ glossaries out there – there are bound to be some good ones. Even that requires searching them out, reading thru them or at least skimming to check quality, etc. How about you see if you can find a few and then provide links to Anthony in the tips & notes section? Obviously I can’t guarantee that he’d use them, but I would suspect he’d at least seriously consider it. Lord I speak REALLY tongue in cheek here because I haven’t checked myself to see if perhaps he’s already provided some links like that on the homepage!!
Next, on the issue of articles being ‘too’ scientific and needing to include more explanation of the basic science. I feel for you there also. Even tho I am a scientist, I work in industry rather than academia – and I’m not a climate scientist, and since graduate school I rarely need to use more than simple basic statistical tests and concepts. Not to mention that the ‘jargon’ varies a bit from one discipline to another. So often there are issues in articles that I’d like to see elaborated on more also – be it to jog my own recollection of the pertinent statistics or because its something I haven’t run across before or what have you. In this article for example, I would very much like to have seen the autocorrelation issue expounded on with just a few more sentences or a paragraph. I posted to that effect and hoped to see either Mr. Eschenbach take the issue up with an update to the article, or someone else perhaps chime in here in the comments that way…. no luck so far. :0) Who knows, perhaps someone will note my second mention of it here and expound a little for me in a comment yet.
Here again there are other aspects that one has to consider, however. First, if the author(s) change the audience aim, e.g., try to write more explanation of the basic science in their articles, then either those articles get both longer (perhaps VERY much longer) or they aren’t able to cover nearly as much ground, aren’t able to address as many different issues as they do. You would probably also wind up with less scientists, and especially less climate scientists, willing to take the time to wade through so much ‘basic’ stuff (to them), in order to read the articles – and if they don’t read them, they’re not going to participate and comment on them either. That would be a great loss to this site. As it is there is often quite a bit in articles that someone well versed in climate science would find to be pretty well known and simplistic…. So its very much a fine line – what is too much, what is too little – both in terms of the technical level in the articles and the length of the articles?
Finally I’d like to address your comment about an objective of modern science being to intimidate anyone who’s not in the good ol’ boys club. Being human beings too, I’m certain that you are right and there are a few, perhaps even some, scientists who behave in that fashion either purposefully or subconsciously. And for all I know, in the academic circles that may have gotten worse in recent years, but I don’t know.
What I can say for certain, however, is that there is a very deliberate and GOOD purpose for the jargon used by scientists that is not in ANY way intended to belittle or awe non-scientists. That purpose is to accurately describe exactly what they are referring to without having to write an entire book. Science, to some extent, is its own language that has developed over time – single words have meanings that often cannot be otherwise expressed without using one or even several sentences. It gets even more complicated because at times a word used in one discipline carries a different meaning than the same word used in another discipline. Regardless, phrases that sound so intimidating are often used in place of what would otherwise require far far longer explanations.
Now any good scientist, who ALSO happens to be a good communicator in layman’s terms, can take almost any scientific word or phrase and make it understandable in basic terms to most people – with a little time and effort, and of course, sometimes far easier than others depending on the topic and the layman’s level of knowledge too. You can have excellent scientists who don’t also happen to be great at translating into lay terms, however. There really is an art to it and its quite often not an easy thing to do.
There are, however, even single words that would take years of study for someone to fully understand all of the meaning and implications inherent in that one word (or phrase). For example, and forgive me, this probably isn’t the best example, but I could use the word osteomyelitis and that will sound utterly foreign and perhaps scary to a layman. I could then tell that layman that it basically means inflammation in bone. The layman might exclaim, ‘why didn’t you just say so!’ and assume that the ‘big word’ was used only to try to hide simple things and make the user appear superior and mysterious. What would be entirely missed is that a doctor or scientist in the field(s) dealing with osteomyelitis, when seeing the word, also thinks/knows of the entire process involved in osteomyelitis including osteoblasts, osteoclasts, various inflammatory cells, enzymes and proteins, the various known causes, treatments, how the surrounding tissue is likely to be involved or affected, etc. So the single word carries a lot more meaning to someone using it and in the discipline reading it than just the simple view that ‘inflamed bone’ would carry to the layman – because the term carries with it the various processes involved in the development, typical course of progress, treatments, etc.
Or there are plenty of terms which, to explain, would require a domino effect of explanations of other key concepts… Who has the time and ability to teach an entire book or course of study in order to put each word or phrase into its full context? That’s why in order for one to become a good scientist, years of guided study are involved, in part so one learns the basic principals and precepts of science itself (e.g., how to properly use the tool that science is), and so one learns the meanings of the various words used within that discipline, and so on.
Its like a mechanic calling for ‘channel locks’ rather than describing the configuration, shape, typical use, and so on of the type of wrench that got named channel lock wrenches… or the carpenter referring to ‘dry wall’ instead of the ‘processed wood product in an 8′ x 4′ x 1/2″ configuration typically used to build walls inside buildings that are subsequently painted or wall papered over to give a finished look to the inside of the structure.’ The mechanic and carpenter aren’t trying to be arrogant and mysterious to inflate their standing and block access to laymen by using terminology involved in their craft, any more than a scientist is. The terminology used within any discipline is just more relevant, accurate, meaningful, and concise than trying to put things into lay terms. We’re just often more familiar with some of the carpenter or mechanic’s terms than we might be with scientific terms, and so we take them for granted.
In other words, some of what may easily be seen as arrogance, condescension, or an attempt to inflate ones position actually has nothing to do with those emotional aspects what-so-ever and are simply how to accurately communicate ones meaning within that field in a reasonably concise way.
Anyhow, for whatever its worth I hope this brief look as some of the other aspects involved in the issues you brought up helps a little.
Lu says:
August 12, 2010 at 7:52 am
Lu, thank you for your comment. I make every effort to make my ideas understandable to most readers, and I am sorry you found them less than clear. If you would point out the wording or the concepts that were difficult to understand, it would help me write better. I, like you, think it is important that everyone understands the issues.
Marty Singh says:
August 12, 2010 at 8:16 am
Not so. The paper says:
So they are not using weather records from “15-20km of the farm”. Nor are they using the “weather variables” directly. Instead, and foolishly in my opinion, they are using trends in the variables …
w.
Hey everyone. Unbelievable. The Beeb has amended their article on this!
“Correction 12th August: this story has been amended to reflect the fact that it is the rate of growth in yields that has fallen, not the yields themselves.”
http://www.bbc.co.uk/news/science-environment-10918591
Willis: I suggest that you read the paper more carefully. The authors first determined the effect of changes in Tmin and Tmax on yield, presumably from monthly or daily temperature data at stations near there sites. The abstract reads: “The farm-level detail, observed over multiple growing seasons, enabled us to construct farm-specific weather variables…” AFTER they had determined that yield dropped X% for every 1 degC increase in local Tmin and improved Y% for every 1 degC increase in local Tmax, the authors may have wanted to put their results in context: How fast is GW causing Tmin and Tmax to change? They obviously couldn’t establish a realistic trend in Tmin and Tmax from a few years of local temperature data, so they calculated trends from 25 years of data (about 0.05 degC/year). From long-term temperature trends, they may have estimated that yields have been dropping X% every 20 years due to effects from Tmin and improving Y% every 20 years due to effects from Tmax (while changing for many other reasons too.)
How do I know this interpretation is correct? I don’t. However, they found that increasing Tmax improved yields and increasing Tmin reduced yields. Over 25 years, both Tmin and Tmax have probably increased. There should be no way to determine that Tmin and Tmax have opposite effects on yield if both have been calculated from long term linear trends! Furthermore, Table S3 in the Supplementary Material also shows that there are different amounts of correlation between Tmin and Tmax during different times in the growing season. The correlation would have to be 1.0 if they were both increasing linearly and -1.0 if Tmin was increasing linearly and Tmax was decreasing linearly.
It is far easier to determine the effect of changes in Tmin from greenhouse studies than from field studies. In the greenhouse, you can keep all of the variables in the greenhouse constant except for Tmin and vary Tmin to widely as desired. In the field, you get whatever Tmin Mother Nature gives you (without as much dynamic range as you’d prefer). The Tmin you do get isn’t a constant in the field, it probably varies 5-10 degC from day to day while the average Tmin (the number you put in the model) varies only 1-2 degC from season to season. The changes in Tmin that Nature provides come along with changes in Tmax, radiation, insects, nutrients, water, rice strain, pH, fungal infection, wind, heavy rainfall, humidity, planting density, seedling health, etc. If you are lucky and careful about what data you include, you may be able to develop a model that predicts how important the key variables are to yield. Unfortunately, this isn’t possible when changes in some variables tend to be correlated with changes in other variables. (Tmin may correlate with cloudiness because clouds block radiative cooling at night. Which one is the cause of poor yield?) Greenhouse experiments can go wrong, too. Maybe Tmin has an effect only when nutrients are limited and you chose to do your experiment in rich soil. However, it is much easier to get a misleading answer in the field than in the greenhouse.
Willis,
Frank is correct. The quote you are referring to is them using their statistical model and applying it to the long-term trend.
Here is what they say about constructing the regression model:
“Our general approach was to regress yield on weather variables and in some specifications, exogenously determined economic variables, whose inclusion improved the precision of the estimated weather impacts. IRRI and its partners
collected data on crop establishment and harvest dates, production inputs, and yields for each farm in each season of each year. They also collected daily weather data from a single monitoring station at each site, which was within 15–20 km of nearly all
farms at a site. This detail enabled us to construct farm-specific measures of weather variables defined according to the rice plant’s three growth phases (for each phase, means for Tmin, Tmax, and radiation and sums for rainfall).”
It just does not make sense to regress the a 26 yr temperature record onto a 6 year yield record. Not that it is just scientifically unsound, but you just can’t mathematically regress vectors of different length against eachother.
Frank – One other reason they stated for doing a field experiment rather than a controlled one is that farmers can correct for changing conditions by changing their fertiliser use etc. Thus this might offset some of the effect of the changes in temperature. Controlled experiments thus mught imply a higher sensitivity than in the real world.
Frank and Marty, thanks for the clarification. You are correct and I was wrong, they did two analyses, one using local data, and one using trends from larger-scale data.
I’m still not impressed by their methods. First, they don’t report (near as I can tell) whether their locally measured climate variables agree with the regional level climate variables or not. If they don’t agree, then looking at the trends is meaningless.
Second, they have not made their data available, so there is no way to determine if they have done the math correctly.
Third, they have not even mentioned the question of autocorrelation. Since temperature data is generally autocorrelated, and yield data is generally very autocorrelated, and their analysis is based upon the statistical significance of the various regression calculations, and statistical significance depends sensitively on autocorrelation, this is a serious omission.
As a result, I see this as a work in progress, without a single verifiable fact in it. As I said, this is too bad, because it is an interesting analysis.
Marty: I still think it is idiotic to look for effects in uncontrolled fields that you have demonstrated occur in the greenhouse and even crazier for granting institutions to invest money under these circumstances. First, find out where the optimum temperature is and how fast yields drop with changes in Tmin and Tmax. If one expects significant changes in yield in the field from changes in Tmin, then look for them. Tmax may be correlated with radiation, explaining why higher Tmax improves yields.
Willis: Monthly temperature data is autocorrelated, but I don’t think annual temperature is. (I think Santer founded that a year’s worth of data provided about 3-4 degrees of freedom, not 12.) If all of the temperature data for one season is used to predict yield, there may not be a problem.